Determining common phase error

Abstract

Systems and techniques for determining common phase error (e.g., phase error common to subcarriers) in an OFDM symbol are described. A respective common phase error for each data symbol of a plurality of data symbols in a data packet are combined to form a refined common phase error. The refined common phase error is based on pilot tones from a plurality of data symbols in a data packet, rather than being limited to pilot tones of a single data packet. Accordingly, refined common phase errors are determined that provide processing gain to accurately estimate common phase error in an OFDM symbol when other systems fail, including when the number of pilot tones in OFDM symbols is low or the data packets are received with high noise. Further improvement is made when the data packet is long due to the coherency of oscillator phase noise and amplifier gain drift.

What is claimed is: 
     
       1. A method of determining phase error, the method comprising:
 receiving a data packet including a training field and a plurality of data symbols, each data symbol of the plurality of data symbols including multiple pilot tones and multiple subcarriers containing data, respectively; 
 obtaining, for each data symbol of the plurality of data symbols, a common phase error based on the multiple pilot tones of the data symbol; 
 combining, for at least one data symbol of the plurality of data symbols, the common phase error for the at least one data symbol with respective ones of the common phase errors for preceding data symbols in the data packet to form a refined phase error, the preceding data symbols preceding the at least one data symbol in the data packet; and 
 determining at least some of the data on the multiple subcarriers of the at least one data symbol based on the refined phase error. 
 
     
     
       2. The method as recited in  claim 1 , wherein obtaining the common phase error for each data symbol includes summing multiple terms for the multiple pilot tones in the data symbol for which the common phase error is obtained. 
     
     
       3. The method as recited in  claim 2 , wherein each of the multiple terms includes an expected data value for the pilot tone, a received data value obtained from the data packet for the pilot tone, and a channel gain value corresponding to a frequency of the pilot tone. 
     
     
       4. The method as recited in  claim 1 , wherein the data packet is received on multiple antennas forming multiple received signals, and the common phase error for each data symbol of the plurality of data symbols is obtained based on the multiple received signals. 
     
     
       5. The method as recited in  claim 1 , wherein the data packet is received on multiple antennas forming multiple received signals, and the common phase error for each data symbol of the plurality of data symbols includes a phase error value obtained separately for each of the multiple received signals. 
     
     
       6. The method as recited in  claim 1 , wherein the data packet is received in multiple space-time encoded signals, and the common phase error for each data symbol of the plurality of data symbols is obtained based on the multiple space-time encoded signals. 
     
     
       7. The method as recited in  claim 1 , wherein the data packet is received in multiple space-time encoded signals, and the common phase error for each data symbol of the plurality of data symbols includes a phase error value obtained separately for each of the multiple space-time encoded signals. 
     
     
       8. The method as recited in  claim 1 , wherein the combining includes, for each data symbol of the preceding data symbols and the at least one data symbol, summing with an infinite-impulse response (IIR) filter:
 a weighted version of the respective common phase error for the data symbol; and 
 a weighted and delayed version of an output of the IIR filter to provide an output of the IIR filter. 
 
     
     
       9. The method as recited in  claim 1 , wherein the refined phase error is formed by applying the common phase error for the preceding data symbols and the common phase error for the at least one data symbol to a leaky integration circuit, and a weight of the leaky integration circuit is adjusted to emphasize the common phase error for a first of the preceding data symbols compared to the common phase error for a second of the preceding data symbols, the first preceding data symbol being closer in the data packet to the at least one data symbol than the second preceding data symbol. 
     
     
       10. The method as recited in  claim 1 , further comprising clearing contributions of at least some of the preceding data symbols in the combining responsive to (i) receiving an indication that a channel gain has changed or (ii) determining that the channel gain has changed. 
     
     
       11. The method as recited in  claim 1 , wherein the combining further includes summing weighted versions of the common phase error for the at least one data symbol and the common phase errors for the preceding data symbols, the weighted versions of the common phase errors determined with weights selected based on a respective position of the data symbol in the data packet. 
     
     
       12. A system for determining phase error, the system comprising:
 a receiver configured to receive a data packet and form a received data packet that includes a training field and a plurality of data symbols, each data symbol of the plurality of data symbols including multiple subcarriers carrying payload data and multiple pilot tones carrying known data values, respectively; 
 a single-symbol phase error module configured to obtain, for each data symbol of the plurality of data symbols, a common phase error for the data symbol based on the known data values and received data values in the received data packet corresponding to the multiple pilot tones; 
 a multi-symbol phase error module configured to combine the common phase error for at least one of the data symbols with respective ones of the common phase errors for preceding data symbols in the received data packet to form a refined phase error by summing weighted versions of the common phase error for the at least one data symbol and the respective common phase errors for the preceding data symbols, the preceding data symbols preceding the at least one data symbol in the received data packet; and 
 a data recovery module configured to determine the payload data of the multiple subcarriers of the at least one data symbol based on the refined phase error. 
 
     
     
       13. The system as recited in  claim 12 , wherein the multi-symbol phase error module includes an infinite-impulse response (IIR) filter that, for each one of the common phase errors:
 sums a weighted version of the common phase error with a weighted version of an output value of a delay element of the IIR filter to form a filter output value; and 
 clocks the filter output value into the delay element; 
 wherein the refined phase error is determined from the filter output value after each respective one of the common phase errors is processed by the IIR filter. 
 
     
     
       14. The system as recited in  claim 13 , wherein the multi-symbol phase error module is further configured to clear contributions of some data symbols by resetting the delay element of the IIR filter to contain a zero value, the resetting responsive to receiving an indication that a training field has been processed. 
     
     
       15. The system as recited in  claim 12 , wherein the weighted versions of common phase error for the at least one data symbol and the common phase errors of the preceding data symbols are determined with weights selected responsive to a level of performance determined for the data packet dropping below a threshold level of performance. 
     
     
       16. The system as recited in  claim 12 , wherein:
 the common phase error for each data symbol is obtained by summing terms for each one of the multiple pilot tones in the data symbol; and 
 each one of the terms for a respective one of the multiple pilot tones includes one of the known data values, a received data value obtained from the data packet for the pilot tone, and a channel gain value corresponding to a frequency of the pilot tone. 
 
     
     
       17. A device for determining phase error, the device comprising:
 a receiver configured to receive a data packet and form a received data packet including a training field and a plurality of data symbols, each data symbol of the plurality of data symbols including multiple data-bearing subcarriers and multiple pilot tones, respectively; 
 a single-symbol phase error module configured to obtain, for each data symbol of the plurality of data symbols, a common phase error based on the multiple pilot tones of the data symbol for which the common phase error is obtained; 
 a multi-symbol phase error module configured to combine the common phase error for each of the data symbols to form a refined phase error by sequentially applying the common phase errors that correspond to each of the data symbols to an input port of an infinite-impulse response (IIR) filter; and 
 a data recovery module configured to determine data on the multiple data-bearing subcarriers based on the refined phase error formed for the plurality of data symbols. 
 
     
     
       18. The device as recited in  claim 17 , wherein a pole location of the IIR filter is adapted when processing one of the data symbols based on a location of the one of the data symbols within the received data packet. 
     
     
       19. The device as recited in  claim 17 , wherein the data packet is received on multiple antennas to form multiple received signals, and the common phase error for each data symbol of the plurality of data symbols is obtained based on values of the multiple pilot tones obtained in each of the multiple received signals. 
     
     
       20. The device as recited in  claim 17 , wherein the data packet is received in multiple space-time encoded signals, and the common phase error for each data symbol of the plurality of data symbols is obtained based on values of the multiple pilot tones obtained in each of the multiple space-time encoded signals.